Transfusion Associated Necrotizing Enterocolitis (Tanec) in Preterm Neonates

Abstract

Background: Etiology of necrotizing enterocolitis (NEC) in preterm infants is multi-factorial. Recent observational studies suggest packed red blood cell (PRBC) transfusion may play a role in NEC. Suggested mechanisms include reperfusion injury or presence of antibodies on donor RBC against intestinal mucosa. Objective: To evaluate the association between exposure to PRBC transfusions and NEC in a large Canadian population-based neonatal database. Methods: A case-control study (1:3) was conducted from neonates admitted to participating sites in the Canadian Neonatal Network between 2003 and 2008. Cases were matched by gestational age. TANEC was defined as exposure to PRBC on the 2 days prior to the day of NEC diagnosis. Controls were compared for the receipt of transfusion within 2 days of discharge. Descriptive statistical methods, Cochran-Armitage trend test, uni- variate conditional logistic regression followed by multiple conditional logistic regression model were employed to control for potential confounders. Results: A total of 927 infants with NEC were matched with randomly selected 2781 controls taken from 5338 patients without NEC. More NEC patients had Apgar <7 at 5 min (28% vs 22%, p=0.0003), higher SNAPII scores (27% vs 23%, p=0.003), were outborn (32% vs 24%, p<0.0001) and had lower birth weight (1298g vs 1335g mean, p=0.001). NEC patients and controls were similar in regards to sex (54.7% vs 55.7% male) and prenatal steroids (74.5% both). Of all patients with NEC, only 87 (9.4%) met TANEC criteria. Exposure to PRBC 2 days before discharge in controls occurred in 5% of cases. The OR for NEC patients with recent PRBC exposure was 2.0 (95%CI 1.48-2.71). The rate of NEC decreased significantly in respect to days after last transfusion (p<0.0001). When controlled for Apgar score, SNAP score, inborn status, birth weight and prenatal steroids, TANEC infants had a higher incidence of PDA (59% vs 33%, p<0.0001), and IVH grade III (38% vs 17%, p<0.0001) than NEC infants without recent PRBC exposure. DISCUSSION: NEC was preceded by PRBC transfusion in the previous 2 days in 9% of cases. Neonatal morbidities in infants with TANEC were increased.